Clouds are significant in the global radiation budget, atmospheric circulation, and hydrological cycle. However, knowledge regarding the observed climatology of the cloud vertical structure (CVS) over Beijing is still poor. Based on high-resolution radiosonde observations at Beijing Nanjiao Weather Observatory (BNWO) during the period 2010–2017, the method for identifying CVS depending on height-resolved relative humidity thresholds is improved, and CVS estimation by radiosonde is compared with observations by millimeter-wave cloud radar and ceilometer at the same site. Good consistency is shown between the three instruments. Then, the CVS climatology, including the frequency distribution and seasonal variation, is investigated. Overall, the occurrence frequency (OF) of cloudy cases in Beijing is slightly higher than that of clear-sky cases, and the cloud OF is highest in summer and lowest in winter. Single-layer clouds and middle-level clouds are dominant in Beijing. In addition, the average cloud top height (CTH), cloud base height (CBH), and cloud thickness in Beijing are 6.2 km, 4.0 km, and 2.2 km, respectively, and show the trend of reaching peaks in spring and minimums in winter. In terms of frequency distribution, the CTH basically resides below an altitude of 16 km, and approximately 43% of the CBHs are located at altitudes of 0.5–1.5 km. The cloud OF has only one peak located at altitudes of 4–8 km in spring, whereas it shows a trimodal distribution in other seasons. The height at which the cloud OF reaches its peak is highest in summer and lowest in winter. To the best of our knowledge, the cloud properties analyzed here are the first to elucidate the distribution and temporal variation of the CVS in Beijing from a long-term sounding perspective, and these results will provide a scientific observation basis for improving the atmospheric circulation model, as well as comparisons and verifications for measurements by ground-based remote sensing equipment.

In satellite health management, anomalies are mostly resolved after an event and are rarely predicted in advance. Thus, trend prediction is critical for avoiding satellite faults, which may affect the accuracy and quality of satellite data and even greatly impact safety. However, it is difficult to predict satellite operation using a simple model because satellite systems are complex and telemetry data are copious, coupled, and intermittent. Therefore, this study proposes a model that combines an attention mechanism and bidirectional long short-term memory (attention-BiLSTM) with telemetry correlation to predict satellite behaviour. First, a high-dimensional K-nearest neighbour mutual information method is used to select the related telemetry variables from multiple variables of satellite telemetry data. Next, we propose a new BiLSTM model with an attention mechanism for telemetry prediction. The dataset used in this study was generated and transmitted from the FY3E meteorological satellite power system. The proposed method was compared with other methods using the same dataset used in the experiment to verify its superiority. The results confirmed that the proposed method outperformed the other methods owing to its prediction precision and superior accuracy, indicating its potential for application in intelligent satellite health management systems.

Satellite housekeeping systems combine heterogeneous sensor channels that require continuous fault monitoring. Existing methods rely mainly on hand-crafted statistics or generic raw-signal models, which can obscure localized morphology and do not explicitly adapt shared features to sensor type. They also rarely assess accuracy together with resource cost and temporal evidence, limiting deployment relevance. To address these limitations, we present OpsSatNet, a lightweight sensor-aware attention network for segment-level anomaly detection from raw, resampled telemetry. Specifically, a dilated residual convolutional backbone captures multi-scale patterns, the feature-wise linear modulation adapts shared features by sensor type, and additive attention pooling aggregates temporal features with learned weights. On OPSSAT-AD, the base model achieves an F1-score of $0.960\pm 0.009$ over five runs, compared with $0.946$ for the strongest published baseline and $0.949\pm 0.004$ for the strongest retrained raw-signal baseline. It contains 258K parameters, while a 16K parameter variant retains an F1-score of $0.952\pm 0.010$ at 8 MFLOPs and 1.157 ms single-thread CPU latency per segment. Ablations show positive mean F1 differences for sensor conditioning and attention pooling, although neither individual effect is statistically significant. These results support lightweight, sensor-aware monitoring, while cross-mission and flight-hardware validation remain necessary.

Accurate typhoon monitoring and tracking are indispensable for disaster mitigation in coastal regions. However, existing methods are hindered by significant spatio-temporal limitations: in the spatial domain, conventional detectors struggle to distinguish weak typhoon vortices amidst severe cloud noise interference; in the temporal domain, standard tracking algorithms often suffer from target loss due to the highly nonlinear and erratic motion of typhoons. To address these challenges, we propose an integrated spatio-temporal framework that combines YOLO11s-ASFF for typhoon detection with DeepSORT for trajectory tracking. YOLO11s provides a lightweight, anchor-free detection architecture, with an adaptive spatial feature fusion (ASFF) stage enhancing multi-scale feature aggregation across detection heads while preserving the pretrained mapping. DeepSORT associates detections using motion-based Kalman filtering and appearance-independent matching, which is suitable for the single-object typhoon tracking scenario. Experiments are conducted on FY-4B Channel13 full-disk satellite imagery using an event-level split, with four storms for training, one for validation, and two for testing. On the held-out Lan and Saola storms, YOLO11s-ASFF achieves F1 0.782 at confidence 0.4, with precision 0.927 and recall 0.677, compared with 0.694 for official YOLO11s. With those detections, DeepSORT achieves MOTA 63.9%. The median center error of true-positive ASFF detections is 56.6 km against interpolated CMA positions, close to 56.3 km for the annotated windows. These results demonstrate the effectiveness of the proposed framework for robust typhoon monitoring and tracking from geostationary satellite imagery.

Novelty detection is the identification (normal or anomalous) of test data differing from the training data (all are normal). It has gained significant attention in many fields, due to the infrequency and unpredictability of anomalies in reality. Most novelty detection methods focus on improving performance under single modal data source conditions. However, in complex realities, relying on single modal data often fails to capture the true state of the object under observation. Thus, these methods cannot always offer accurate and reliable decisions in novelty detection. Moreover, research on novelty detection under multi-modal conditions is relatively limited. Therefore, we propose MSENet, an end-to-end multi-modal semantic-enhanced network based on autoencoder for novelty detection, including the feature-extraction and semantic-mining modules, which can effectively reconstruct the multi-modal data belonging to a target class by bridging the semantic gaps between different modalities. In addition, we develop an L₁^* regularization method based on L₁-norm to select more representative features by promoting sparser weights during the optimization process. The results of the novelty detection experiment on two multi-modal datasets demonstrate that our method outperforms state-of-the-art methods. Furthermore, an ablation study and a regularization experiment verified the effectiveness of the MSENet and L₁^* regularization methods.

At present, satellite anomaly is mostly solved after the event, and rarely predicted in advance in satellite health management. Thus, satellite trend prediction is quietly important for avoiding the fault which perhaps affects data accuracy and service quality of satellite, and even impacts greatly on satellite safety. However, it is difficult to predict satellite operation through a simple model because satellite system is complex, and telemetry data is numerous, coupled and spatiotemporal. Therefore, this paper proposes a model combing attention mechanism and Bidirectional Long Short-term Memory (Attention-BiLSTM) with correlation telemetry to predict the situation of satellite operation. Firstly, high-dimensional K-NearestNeighbor Mutual Information (HKNN-MI) method is performed to select the related telemetry variables from multiple variables of satellite telemetry data. Secondly, we put forward to a new BiLSTM model with attention mechanism for telemetry prediction. The dataset for the research is generated and transmitted from the power system of FY3E meteorological satellite. In order to verify the superiority of the proposed model, it is compared with other method based on the same dataset in the experiment. The result shows that the method outperforms other methods due to its better accuracy and prediction precision.